arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.07035cs.ROcs.LGcs.SYeess.SY
收藏

用于自动驾驶赛车车辆控制的残差策略学习

Residual Policy Learning for Vehicle Control of Autonomous Racing Cars

Technical University of Munich(慕尼黑工业大学)

Raphael Trumpp, Denis Hoornaert, Marco Caccamo

AI总结:

本文提出一种用于自动驾驶赛车的残差车辆控制器,通过残差策略学习修正经典路径跟随控制器,在F1TENTH仿真环境及十二条真实赛道评估中平均将单圈时间缩短4.55%,并能在未知赛道上获得提升。

摘要:

自动驾驶赛车车辆控制器的开发具有挑战性,因为赛车在其物理驾驶极限下运行。受性能提升需求的推动,自动驾驶赛车研究中基于机器学习的控制器大量涌现。尽管这些方法表现出有竞争力的性能,但其实际适用性往往受到限制。残差策略学习有望通过将经典控制器与学习到的残差控制器相结合来缓解这一缺陷。残差控制器的关键优势在于其高适应性,同时兼具经典控制器的稳定行为。我们提出了一种用于自动驾驶赛车的残差车辆控制器,该控制器学习修正经典控制器以实现赛车路线的路径跟随。在一项广泛研究中,我们针对F1TENTH自动驾驶赛车系列的仿真车辆评估了该方法的性能提升。对十二条复现的真实世界赛道的评估表明,与经典控制器相比,残差控制器将单圈时间平均缩短了4.55%,甚至在未知赛道上也能实现单圈时间的提升。

英文摘要:

The development of vehicle controllers for autonomous racing is challenging because racing cars operate at their physical driving limit. Prompted by the demand for improved performance, autonomous racing research has seen the proliferation of machine learning-based controllers. While these approaches show competitive performance, their practical applicability is often limited. Residual policy learning promises to mitigate this drawback by combining classical controllers with learned residual controllers. The critical advantage of residual controllers is their high adaptability parallel to the classical controller's stable behavior. We propose a residual vehicle controller for autonomous racing cars that learns to amend a classical controller for the path-following of racing lines. In an extensive study, performance gains of our approach are evaluated for a simulated car of the F1TENTH autonomous racing series. The evaluation for twelve replicated real-world racetracks shows that the residual controller reduces lap times by an average of 4.55 % compared to a classical controller and even enables lap time gains on unknown racetracks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑